跳到正文
研踪PaperTrail

论文库

检索计算机领域论文,按研究方向、作者或关键词筛选。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+816新增论文
+642其中 AI 相关
+95已有论文修订
+95新增 AI 导读

arXiv 最近完成扫描:2026-11-21 12:19 · 全站最近入库:2026-11-21 12:19各信源进度 →

统计口径与数据新鲜度

统计 2026-11-20 14:37 — 2026-11-21 14:37(北京时间) · 点击数字查看对应简报;AI 相关包含在新增论文内。

最近完成检查:2026-11-21 12:19最近发现新增版本:2026-11-21 12:19库内论文最新更新时间:2026-11-20 02:59

每分钟刷新已保存的数据。空检查不代表采集异常;新收录可能包含历史论文。AI 导读按新增版本计数。

29,158 已收录论文24,181 AI 相关8,992 有修订112 HF 社区发现

浏览论文库

时间范围跟随排序:版本更新、本站收录或首次发表。最新收录可能包含历史论文。部分信源异常,查看状态 →

找到 6,976 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-21 12:17(北京时间)
新论文 · v1arXiv:2611.23096

Hidden-State Similarity Predicts Re-Elicitation After Inoculation Prompting

Rose N. Kwon, Samy Mammeri, Sabyasachi Sahoo, Christian Gagné, Thomas Jiralerspong

原文摘要

Fine-tuning on narrow harmful tasks can cause emergent misalignment, where models generalize harmful behavior beyond the training distribution. Inoculation prompting can reduce this effect by explicitly eliciting the undesired behavior during training, but recent work shows that the behavior can reappear when evaluation prompts contain cues from the training context. We study what makes such prompts effective triggers. We find that, surprisingly, textual similarity to the inoculation prompt is an incomplete predictor. Instead, prompts are more likely to re-elicit suppressed behavior when they induce activation states similar to those produced by the inoculation context. These findings advance our understanding of how inoculation prompting modulates conditional misalignment, and suggest that activation space analysis can help identify when suppressed behaviors remain accessible under eval-time prompts.

cs.AI

版本更新 · v2arXiv:2608.27507

Learning Complementary Routes for Scouting Persimmon Orchards during Shoot Growth with CoverShare

Junhao Cao, Tao Wang, Zipeng Xie, Hongyi Xia, Jianian Wu, Lixia Huang

原文摘要

At the study orchard, scouting during persimmon shoot growth focuses on leaf color change, leaf area, and internode length. Several routes may share an access lane, yet each can offer a view of row segments that the others miss. We introduce CoverShare to learn complementary routes under a fixed action budget. Within Policy Gradient for Parallel State Entropy maximization (PGPSE), CoverShare redistributes auxiliary rewards according to the state coverage lost when each policy is omitted and the number of policies visiting each state. It preserves the pooled entropy objective and total auxiliary reward. We simulated the annotated road network of a 6.7-ha site with 107 observable persimmon scouting units across five zones. At 40 actions per route, six distinct policies covered an expected 17.84 units perinspection, compared with 8.28 for six independent runs of one policy, averaged across eight training seeds. With the same policies trained on 40-action routes, extending each inspection route to 80 and 120 actions raised coverage to 22.52 and 28.38 units. Across seven public discrete-state tasks, CoverShare increased state support by a mean of 16.29% over the team-entropy baseline. Controlled ablations identified leave-one-policy-out coverage as the main contributor to these gains.

cs.LG · cs.AI

版本更新 · v2arXiv:2607.07646

RL Post-Training Builds Compositional Reasoning Strategies

Azwar Abdulsalam, Nishil Patel, Andrew Saxe

原文摘要

Does RL post-training merely amplify primitive skills already latent in a base model, or can it compose primitive skills into new higher-level strategies? We study this question in a fully observable rewrite-grammar environment where the pretraining distribution is known and every generated rewrite can be audited. A Transformer is pretrained on primitive symbol-rewrite chains and post-trained on a Trace-based reasoning task with only a binary final-answer reward. RL solves held-out problems that remain rarely solved by the pretrained model even under much larger sampling budgets, while rejection fine-tuning improves early but plateaus. Trace analysis shows that RL reorganizes primitive competence through a phased compositional mechanism: it first strengthens primitive reductions, then discovers valid composed procedures. These include sequential compositions, which collapse ordered chains of primitive contractions, and parallel compositions, which combine independent primitive contractions in a single step. The composed procedures are not isolated samples; they are reused and consolidated into a stable repertoire. Comparing RL with rejection fine-tuning shows that the key difference is not exploration volume but selectivity: RFT produces many shortcut-like rewrites, much of them invalid, whereas RL concentrates exploration into valid reusable structure. Pretraining ablations show that the emergence of compositional strategies is gated not by primitive exposure alone, but by whether pretraining organizes primitive competence into reduction procedures that RL can later compress. The base model provides weak procedural ingredients; RL builds them into reliable higher-level strategies.

cs.AI · cs.CL

新论文 · v1arXiv:2611.23093

What's in a Tree ? Hierarchical $n$-gram models as controlled approximators of language

Aritra Das, Francesco Cagnetta, Darshil Doshi, Victor V. Albert, Maissam Barkeshli, Andrey Gromov

原文摘要

Hierarchical models have been proposed as synthetic settings for studying language. However, their relevance to realistic language corpora remains underexplored. In this work, we introduce tree-grams, autoregressive generative $n$-gram models that build the conditional probability tensor $p(x_n \mid x_{1:n-1})$ through a hierarchical composition of order-2 and order-3 tensors. $n$ controls the maximal context window, whereas a rank hyperparameter $d$ controls model capacity, yielding an interpretable and systematically scalable family of approximators of language. We show how tree-grams outperform existing $n$-gram baselines, while remaining competitive with deep transformer-based language models. We demonstrate the utility of such controlled approximations by using tree-grams as generative models. In particular, we show that hyperparameter transfer under $μ$P can be made more robust when narrower models are tuned on simpler data, generated by tree-grams with moderate $n$ and $d$. In addition, we show that entropy and effective range of correlations control performance gap after fine-tuning. This is because tree-grams allow us to explicitly control horizon and capacity. When allowed to route across candidate topologies, we observe emergent sparsity in tree-grams. Such concentrated routing permits pruning upto 95\% of topologies without significant change in performance.

cs.CL · cond-mat.dis-nn · cs.AI · cs.LG

版本更新 · v2arXiv:2611.14808

OverServe: Resource Amplification Attacks on SaaS Agents via Tool--Prompt Co-Design

Ruiheng Yu, Oteo Mamo, Siyu Gan, Weikuan Yu, Mulong Luo

原文摘要

Advances in large language models (LLMs) are enabling software-as-a-service (SaaS) providers to automate longer business workflows with agents. In subscription-based SaaS deployments, model and tool calls consume provider resources without additional customer payments. Because the agent autonomously decides when to stop, a malicious tenant can send a prompt to the agent, exploiting its completion decisions to extend execution and waste provider resources. In this paper, we present OverServe, a multi-agent framework for constructing such attacks against SaaS agents. OverServe co-designs Model Context Protocol (MCP) status tools and completion-confirmation prompts for the attacker. Each tool call returns accurate status and terminates normally, yet the agent may repeatedly check an unfinished condition to satisfy the prompt. This consumes the SaaS provider's resources even after the original business task is completed. OverServe constructs service-specific tool--prompt pairs and uses separate tool and prompt reviews to guide candidate pair selection. Tested tool specifications trigger no alerts in the evaluated MCP scanners. We evaluate Qwen, Nemotron, DeepSeek, and Sonnet on selected tasks from nine benchmarks. All OverServe runs pass the original task checks. OverServe can reach a maximal amplification of API requests by $93.88\times$ while preserving the original task goal and collateral checks.

cs.CR · cs.AI · cs.SE

新论文 · v1arXiv:2611.23091

BugGen 2.0: Rethinking the Roles of LLMs for Scalable RTL Bug Synthesis

Minh Luu, Michael Quinn, Aakash Tyagi, David Houngninou, Jiang Hu

原文摘要

AI-assisted RTL localization and debugging need ground-truth bug datasets with known locations and observable failures, while testbench grading uses injected bugs to measure detection capability. Creating such datasets at scale remains a major EDA challenge. BugGen 1.0 shows that Large Language Models (LLMs) can generate context-aware RTL mutations. Still, it relies on the model for both choosing where to mutate and generating the edit, which increases inference costs and can lead to invalid choices. BugGen 2.0 changes this division of labor. Deterministic algorithms handle structural analysis, target selection, and validation, while the LLM focuses on localized semantic rewriting. Across four OpenTitan IP blocks and FabScalar with three LLMs, BugGen 2.0 reaches every requested quota of compilable bug scenarios that cause at least two test failures, with 14-16 of 16 configured mutation types represented. In completed same-model comparisons, it reduces token use by 6.8x-15.4x relative to BugGen 1.0 and 1.5x-7.9x relative to Firefly. BugGen 2.0 also expands the dataset without simulating every new bug. After finding accepted scenarios with observable failures, it generates alternative mutations at the same locations, simulates only a subset, and estimates acceptance for the rest. This skips 63.6% of AES and 33.9% of FabScalar variant simulations. Full simulation finds that 77.7% of AES variants and 95.2% of FabScalar variants are accepted, with a 1.09% estimation error on the directly comparable FabScalar set. These results show that focusing LLMs on semantic rewriting lowers inference and simulation cost while improving consistency across models and designs.

cs.SE · cs.AI

版本更新 · v2arXiv:2608.11537

Semantic Prism: Generative Segmentation with an Explicit Semantic-Image Interface

Weize Cai, Yongqi Dong, Zhida Shao, Zixin Fu, Xiaoxi Hu

原文摘要

Conditional image generators can render segmentation maps as semantic RGB images, but color drift and boundary mixing can undermine direct decoding. We present Semantic Prism, a framework that combines an explicit semantic-image interface with hierarchical refinement for road-scene segmentation. A generator adapted from a diffusion-distilled backbone produces a semantic RGB image with a single denoising update per crop. Distances to a fixed class-color codebook define interface logits and pixelwise class probabilities. Hierarchical Generator Evidence Alignment (HGEA) combines spatially aligned multilevel features from the VAE encoder of the generator with the input image, generated semantic image, and interface probabilities to predict additive corrections to these logits. Contextual Interface Hierarchy Disagreement (C-IHD) combines pointwise uncertainty, local context, and interface-refinement disagreement in a fixed score for ranking pixel errors, without an auxiliary predictor or an additional network pass. On the 500-image Cityscapes validation set, Semantic Prism achieves 72.07% mean mIoU, improving on direct-interface decoding by 11.39 percentage points. Extensive capacity-matched ablations show higher mean mIoU for multilevel refinement than for flat and single-level alternatives. The Cityscapes-trained model transfers to ACDC Dataset, achieving 46.89% mIoU without target-domain adaptation. Beyond Cityscapes, on BDD100K dataset, a separately trained model achieves 62.22% mIoU on its full 1,000-image validation set. Across the 3 evaluated datasets, C-IHD improves pixel-error area under the AUPR over uncertainty derived from maximum softmax probability, with segmentation predictions held fixed for each comparison. In particular, on ACDC, AUPR increases from 0.6580 to 0.7557, demonstrating that Semantic Prism presents more effective error ranking under adverse-condition domain shift.

cs.CV · cs.AI · cs.LG · eess.IV

新论文 · v1arXiv:2611.23081

Learning to Undo: Transfer Reinforcement Learning under State Space Transformations

Mridul Mahajan, Aldo Pacchiano, Xuezhou Zhang

原文摘要

Transfer learning in reinforcement learning (RL) has shown strong empirical success. In this work, we take a more principled perspective by studying when and how transferring knowledge between MDPs (a source and a target) can be provably beneficial. Specifically, we consider the case where there exists an undo map such that applying this map to the target's state space recovers the source exactly. We propose an algorithm that learns this map by matching state feature statistics gathered from both MDPs, and then uses it to transfer the source policy to the target. We theoretically justify the algorithm by analyzing the setting when the undo map is linear and the source is linearly-$Q^\star$ realizable, where our approach has strictly better sample complexity than tabula rasa RL in the target MDP. Empirically, we demonstrate that these benefits extend beyond this regime: on challenging continuous control tasks and Atari games, our method achieves significantly better sample efficiency. Overall, our results highlight how shared structure between tasks can be leveraged for efficient transfer of policies across environments.

cs.LG · cs.AI

新论文 · v1arXiv:2611.23080

Colombo: A Frontier-to-Frontier Autoresearch System for Oracle-Based Optimization

Haihan Zhang, Wendao Wu, Chenheng Zhang, Chunyuan Zheng, Yanyi Li, Cong Fang, Haoxuan Li, Zhouchen Lin

原文摘要

Science advances from frontier to frontier as discoveries reshape what is known and create the basis for further discovery. We present Colombo, a self-organizing AI researcher that drives this renewal, first instantiated in oracle-based optimization. Here, the central question is how efficiently a problem class can be solved through a prescribed information interface and what costs no admissible algorithm can avoid. The gap between achievable guarantees and unavoidable limits defines the frontier recorded in its Complexity Atlas, a living map of frontier settings and the current state of knowledge. Self-Organization turns unresolved gaps into adaptive research programmes, coordinating conjecture formation, mathematical investigation, and replanning as evidence accumulates. Adversarial Co-Ascent develops algorithms and hard families through reciprocal challenge, exposing mechanisms of information use and concealment that guide stronger methods and proofs of unavoidable costs. Independent review and Lean-based formal certification check mathematical claims, while the Long-Term Memory carries supported findings and research experience across settings. To date, Colombo has established optimal or near-optimal rates for more than fifty settings in the Atlas. Colombo thus takes a concrete step toward Autopoietic Science: a scientific process whose validated advances regenerate the conditions of its own progress, continually reshaping how inquiry is organized and what it can discover.

cs.AI · math.OC

新论文 · v1arXiv:2611.23074

Runtime Verification of Multiple Natural Language Criteria for Agent Governance

Silviu Pitis, Parand A. Alamdari, Jessica Tang, Toryn Q. Klassen, Sheila A. McIlraith

原文摘要

AI agents are governed by hundreds of natural language criteria, yet existing verifiers cannot efficiently evaluate an agent's output against a large set of such criteria and report which are satisfied, violated, or inapplicable. We introduce the VFM, a tree-attention verifier that encodes the shared context and target text, and scores N natural language criteria in parallel, returning independent per-criterion ternary verdicts. The VFM is pretrained on CriteriaBank, a corpus of 394K (context, target, criterion, verdict) tuples spanning over 321K distinct criterion strings. A finetuned VFM-4B outperforms GPT-5.4 on a synthetic governance benchmark. On the DynaBench multi-criterion benchmark, it outperforms DynaGuard-4B on trace-level failure detection, and its per-criterion verdicts localize the violated rule and support focused escalation to a larger chain-of-thought judge. We show that CriteriaBank pretraining improves cross-domain transfer, particularly in the low-data regime, while matching a strong direct-finetuning control at full data. On a single H100, VFM-4B evaluates 1,000 criteria against a 4K-token context in under 2s at <11 GB VRAM, making it suitable for runtime verification.

cs.CL · cs.AI · cs.LG

新论文 · v1arXiv:2611.23068

The Resolution Hypothesis: Agent Evolution's Key Is Identifying Improvements

Yuanhao Zeng, Shaocheng Lan, Ao Lu, Kuang Yan, Heng Lin, Xinang Xu, Yuan Luo, Yueting Wang 等 11 位作者

原文摘要

Agent evolution follows a simple loop: a proposer writes candidate artifacts (prompts, skills, memories) and selection accepts those that score higher. This loop has shown promise, yet practitioners face a puzzle: methods trade wins across papers, and evolution often fails to bootstrap on certain tasks. We trace this puzzle to the Resolution Hypothesis: agent evolution stalls not because methods fail to propose better artifacts, but because noisy evaluations cannot reliably identify them. Across seven evolution methods and six models (over 10M LLM calls), we find support for the hypothesis at both levels where evaluation scores are used. At the candidate level (choosing artifacts), helpful candidate artifacts exist, but selection is no better than random at telling them from harmful ones, so errors accumulate across rounds. At the method level (ranking methods), typical margins of a few percentage points are smaller than the difference between two independent runs of the same method, so published rankings often rest on luck. Together, these findings indicate that agent evolution is often bottlenecked by reliably identifying improvements. This calls for a shift in how we develop and evaluate self-improving agents, from designing ever more candidate improvement systems to making true improvements detectable and accumulable.

cs.AI · cs.CL

新论文 · v1arXiv:2611.23067

Values Shape What Self-Improving Coding Agents Build

Jason Liang, Dylan Waldner, Risto Miikkulainen

原文摘要

Self-improving agents are typically optimized for the results they produce, with less attention to how they produce them. Their instructions give designers limited ways to shape how they work, especially when the task is open-ended. This paper studies whether a short value statement in an agent's persistent instruction file changes not only how well the agent performs but also what it builds. Four higher-order value types in Schwartz's theory of basic human values were evaluated: openness to change, conservation, self-enhancement, and self-transcendence. Across diverse coding tasks, the values were held fixed within each run over multiple iterations of self-improvement. Three findings emerged: (1) Values matter for performance across a diverse set of tasks. The openness values increase performance on held-out data, whereas the opposite conservation values result in scores below those of an agent given no values. (2) Values change what agents build, producing solution algorithms that differ in size, complexity and approach. (3) Human oversight is still needed for designing values. Allowing the agent to revise its own values narrows the differences between sets of values, and values distilled from the AI literature provide no detectable gain over having no values. Properly designed values can thus influence what a self-improving coding agent builds and, where the task permits, how well it performs.

cs.AI · cs.MA

新论文 · v1arXiv:2611.23063

Deep Reinforcement Learning for Heat Pump and Underfloor Heating Control: An End-to-End Case Study

Antonio Manjavacas, David Suárez-Caro, Alejandro Campoy-Nieves, Guillermo Rey, Gonzalo Satrústegui, Miguel Molina-Solana

原文摘要

Efficient control of underfloor heating (UFH) systems coupled with heat pumps (HPs) is challenging due to the high thermal inertia of buildings and the slow dynamic response of UFH systems. Deep Reinforcement Learning (DRL) offers a data-driven control approach capable of simultaneously optimizing energy consumption and thermal comfort. However, the application of DRL-based controllers in occupied buildings under real operating conditions remains limited. This work presents and end-to-end case study (development, evaluation, and real-world deployment) of a DRL-based controller for an integrated UFH-HP system. First, a simulation-based pilot study was conducted to assess different DRL algorithms, control strategies, and configurations. Subsequently, a digital twin of a real residential building was developed and used to train DRL agents under simulated operating conditions. The best-performing agent was then deployed in the physical building without additional online fine-tuning. The results show that the controller successfully maintained indoor temperatures within the prescribed comfort limits while dynamically adjusting the HP supply temperature and UFH valve positions. These findings demonstrate the feasibility of deploying DRL controllers in occupied buildings and highlight their robustness and the practical challenges associated with transferring control policies from digital environments to real heating systems.

cs.AI · eess.SY

新论文 · v1arXiv:2611.23055

FrameRouter: Frame Budget Routing for Long-Video Understanding on Video-MME-v2 and Beyond

Bingjun Luo, Jialin Guo, Siqi Li

原文摘要

Long-video question answering benchmarks like Video-MME-v2 typically evaluate systems under a uniform frame budget for every question, despite stark variation in the evidence each query actually demands. We analyze this mismatch on Video-MME-v2 and find that different question subsets prefer different frozen frame-budget policies, while uniformly increasing the frame count does not reliably improve all metrics. These findings motivate fixed-mean frame-budget routing, a controlled inference setting in which each question may receive a different frame budget while the average budget matches a uniform reference. We introduce FrameRouter, a training-free test-time framework that combines an evidence-demand router, a budget-constrained allocation rule, and a frozen bank of frame-sampling policies. By controlling how much visual evidence each question receives while leaving the underlying samplers unchanged, FrameRouter is orthogonal to query-aware frame selection. Across Video-MME-v2 and several long-video understanding benchmarks, FrameRouter improves long-video question answering by using the same average visual budget more selectively across questions. Code is available at https://github.com/bingjunluo/FrameRouter.

cs.AI · cs.CV · cs.LG

新论文 · v1arXiv:2611.23054

ReTouchGS: Ultralight Render-free Aesthetic Color Enhancement and Transfer for 3D Gaussian Splatting

Chuanzhi Xu, Boyu Wei, Chengkun Yue, Xuanhua Yin, Yiyang Li, Haodong Chen, Qiang Qu, Weidong Cai

原文摘要

3D Gaussian Splatting (3DGS) enables photorealistic scene reconstruction. Existing 3DGS color transfer and interactive editing methods are guided by reference images and user-specified color adjustments. However, in many applications, users simply want aesthetically guided color enhancement that automatically makes the scene more visually appealing while maintaining consistent colors and tones across viewpoints. We present ReTouchGS, to our knowledge the first framework for reference-free aesthetic color enhancement of pretrained 3DGS models. It predicts a scene-level color lookup table (LUT) from the model and its associated camera parameters. ReTouchGS also supports reference-guided color transfer by fitting a scene-level LUT using one edited view of the same scene and its corresponding original render. A shared blend-aware color refinement network with only 44.7K parameters refines an analytic color update to realize either color mapping without rendering. It preserves geometry and delivers standard 3DGS assets in seconds. We also design the ReTouch3D benchmark to support the evaluation of aesthetic color enhancement and reference-guided color transfer across viewpoints. Experiments demonstrate high color accuracy and strong generalization to unseen scenes in both aesthetic color enhancement and reference-guided color transfer.

cs.CV · cs.AI · cs.GR

新论文 · v1arXiv:2611.23045

ForecastGym: A Live, Controlled, and Replayable Benchmark for Forecasting Agents

Yicheng Tao, Yuante Li, Zipei Fan

原文摘要

Agents built on large language models (LLMs) increasingly forecast real-world events by searching the Web and reasoning over what they find. How well such an agent forecasts depends on both the retrieved evidence and the model, making control over that evidence essential for evaluation. Live evaluations can mix retrieval with reasoning when each system searches on its own, while retrospective search can leak later information through date filters or today's ranking of historical pages. Comparing models, search sources, and memory strategies requires both fixed evidence and a record of how it evolved before resolution. We introduce ForecastGym, which records what four commercial web search APIs return for open prediction-market questions every two days and replays this frozen evidence offline once outcomes are known, comparing models, memory, and captured search sources under a fixed evidence-acquisition policy. Its first release covers 150 questions, 614 evidence states, one per question and collection time, and 131k search results. Rerunning the same queries after settlement returns results that all but state the outcome for up to 92% of the questions, despite the tested date filters. On the frozen evidence, all tested LLMs trail the market in forecasting accuracy. Our diagnostic analysis suggests that missed public information contributes more to the gap in high-gap states than misreading does. Switching the search API changes a model's Brier score by up to 0.043, more than the spread across models, while persistent notes score similarly to rereading all evidence at half the input tokens. The evidence a forecaster sees can matter as much as the model that reads it.

cs.AI · cs.CL

版本更新 · v2arXiv:2607.27951

LLM Misuse Safeguards Rely on Untrusted Input: A Widespread, High-Risk Vulnerability

Pingyu Wu, Weiming Zhang, Nenghai Yu

原文摘要

Large language models now help professionals in fields such as cybersecurity and biology, yet the same answers would help an attacker. Deployers rely on misuse safeguards to keep such answers from attackers. We find that these safeguards share a high-risk vulnerability: they decide whether to release an answer from what users write, including the professional identity or purpose they claim, and anyone can write such a claim. Threat reports from model vendors and security firms show that attackers who fabricate a professional identity or purpose already exploit this vulnerability to cause serious harm. The vulnerability is also widespread across the state of the art: every defense proposed in 129 papers from the latest editions of 12 top venues has it. Safeguard research studies the wrong object: it treats harm as a property of the text of a request, although the same text can come from a professional or an attacker. Current safeguard benchmarks therefore cannot reveal the vulnerability: a safeguard can score high on them and still give dangerous answers to attackers. Because the vulnerability lies in what users write, no improvement in how safeguards judge that writing can close it. Under the patch we propose, large language models can help professionals without giving dangerous capabilities to anyone who merely claims to be one.

cs.CR · cs.AI

版本更新 · v2arXiv:2609.17953

Uncovering Faithfulness Gaps in VLMs via Explanation-Driven Counterfactual Testing

Sihao Ding, Santosh Vasa, Aditi Ramadwar, Thomas Monninger

原文摘要

Vision-Language Models (VLMs) can produce Natural Language Explanations (NLEs) that sound plausible yet remain inconsistent with the visual evidence they cite. We present Explanation-Driven Counterfactual Testing (EDCT), an intervention-based protocol that extracts visual concepts cited in a model's explanation, applies verified minimal edits to them, and tests whether the resulting answer and explanation remain consistent with the edited image. Using this protocol, we create EDCT-Bench, a comprehensive benchmark spanning three complementary domains: knowledge-intensive visual question answering (OK-VQA), safety-critical driving (DriveLM), and 3D spatial reasoning (3DSRBench). Across the evaluated VLMs, EDCT reveals substantial faithfulness gaps, with models frequently producing responses inconsistent with verified visual changes. Finally, our fine-tuning study suggests that EDCT-generated counterfactuals provide high-impact training signals.

cs.CV · cs.AI

新论文 · v1arXiv:2611.23012

PeriaTrace: Tooth-Level Grounding Diagnosis and Patient-Facing Explanations from Periapical Radiographs

Zhenhuan Zhou, Yuchen Zhang, Jingbo Zhu, Shiqing Ma, Yanlin Wu, Tao Li

原文摘要

Periapical radiographs (PRs) are central to endodontic diagnosis, but interpreting them and preparing reports are time-consuming. Technical language can limit patients' understanding, while existing automated methods often lack tooth-level links between findings and image evidence. We present PeriaTrace, a three-stage framework integrating tooth-level diagnosis, visually grounded report generation, and patient-friendly explanations and care recommendations. PeriaTrace was developed and evaluated using 12,877 PRs from an internal dataset and two external clinical cohorts. In internal evaluation, it achieved a macro Dice coefficient of 85.35%, BLEU-4 of 0.6504, an F1 score of 94.37% for joint tooth localization and Fédération Dentaire Internationale (FDI) identification, and a grounding-tag F1 score of 77.93%. In the two external cohorts, senior endodontists considered 91.16% and 84.52% of reports suitable for retention or revision without complete rewriting. Generation averaged 4.41 seconds per PR, several tens of times faster than manual drafting. Among 196 survey participants without medical backgrounds, 96.4% preferred reports with plain-language explanations and care recommendations, and 84.7% found visual localization cues helpful. PeriaTrace may enable faster, evidence-traceable reporting while supporting clinical review and patient communication. (Abridged version)

cs.AI

版本更新 · v2arXiv:2609.33319

PhysAlign: A Benchmark for Evidence-Grounded Role Alignment in Multimodal Physics Reasoning

Kecheng Liang, Haoyang Liu, Zexin Chen, Zirong Liu, Weixing Chen, Qiufeng Wang, Yang Liu, Liang Lin

原文摘要

A key challenge in physics diagram understanding is correctly associating visual information with the physical entities, relations, and conditions it describes. Even when a value, symbol, or other local element is accurately recognized, assigning it to the wrong entity or scope can distort the underlying physical premise and lead to incorrect reasoning. To systematically study this challenge, we introduce \textbf{PhysAlign}, a benchmark designed to assess whether multimodal models correctly associate information recognized from physics diagrams with its intended physical role. By disentangling visual recognition from physical-role assignment through localized probes and controlled variants, PhysAlign isolates correspondence errors from recognition failures. It contains 3,341 human-validated probes spanning 986 physics problems, enabling systematic evaluation of visual recognition and physical-role correspondence at scale. We further introduce five complementary evaluation metrics, including CAcc, GAcc, and JAcc, which provide a comprehensive assessment of models' ability to recognize diagram content, establish correct physical correspondences, and solve the underlying physics problem. Across our evaluated multimodal models, PhysAlign reveals a consistent gap between local visual recognition and physical-role grounding. Even when the queried content is correctly recognized, the conditional correspondence error rate remains 13.8\% for GPT-6-Astra and rises to about 50.6\% for InternVL3.5-8B. These findings indicate that strong perception alone does not ensure reliable physical interpretation, exposing a distinct grounding bottleneck that is largely hidden by answer-level accuracy and highlighting the need for future models to better align recognized visual evidence with its physical meaning.

cs.AI

新论文 · v1arXiv:2611.22998

Residual-Guided Lightweight Reinforcement Learning for Risk-Aware Trajectory Selection with a Frozen World Model

Jiaqi Wu, Yongqi Dong, Yijie Zhong, Maple Shao, Yide Tao, Linsen Cai, Weize Cai, Hai L. Vu 等 9 位作者

原文摘要

In automated driving, world models provide future scene predictions for evaluating candidate trajectories, but fixed selection rules may not capture scene-dependent trade-offs. We propose a residual-guided reinforcement learning (RL) framework for lightweight trajectory selection with a frozen world model and a fixed candidate bank. The framework uses proximal policy optimization (PPO) to fine-tune a joint policy over candidate selection and bounded trajectory residuals. A context-conditioned ranking network learns candidate preferences from feedback obtained by evaluating local perturbations of the selected trajectories. At inference, the selector combines learned preference and predicted risk ranks using fixed weights and returns the selected candidate trajectory without modification. Fine-tuning updates 16,106 parameters across the actor and critic, while the deployed selector retains only the 11,725-parameter ranking network, with the residual branch and critic used exclusively during training. In open-loop evaluation on the nuScenes validation set, the method reduces average Euclidean trajectory error (L2) by 3.99% relative to the RiskWorld model baseline without RL-based selection. Under matched budgets for feedback accesses and actor updates, residual exploration reduces L2 by 0.83% relative to the categorical PPO variant trained without residuals. The method also maintains lower L2 than RiskWorld when risk features are removed or shuffled.

cs.AI · cs.ET · cs.LG · cs.RO · eess.SY

版本更新 · v2arXiv:2610.00620

Misalignment of Low-Loss Regions Causes Grokking

Yongding Tian, Zaid Al-Ars, Maksim Kitsak, Peter Hofstee

原文摘要

Grokking refers to the delayed emergence of validation-set generalization after a model has already overfit the training set. Although first observed in small algorithmic tasks trained with transformers, its underlying mechanism remains unsettled. In this work, we develop an analysis framework based on mode connectivity and the geometry of low-loss regions. The framework predicts that the standard modular-arithmetic setting does not always produce grokking: under a symmetry-preserving train/validation split, we observe a stable anti-grokking case in which validation performance does not recover. This counterexample challenges several existing correlational explanations of grokking. More broadly, our analysis framework and results further suggest that grokking arises when the low-loss regions induced by the training and validation partitions are misaligned. Once these regions become well aligned, training hyperparameters alone cannot produce grokking and the observed dynamics collapse to either trainable or non-trainable behavior.

cs.LG · cs.AI

新论文 · v1arXiv:2611.22991

RagHarnessDebugger: Jev-Guided Agentic RAG in Contract Inference Repair Auditing

Moriya Dechtiar, Daniel Martin Katz, Michael J. Bommarito, Hongming Wang, Sylvain Jaume

原文摘要

Agentic systems using retrieval-augmented generation (RAG) must repair their own answers by gathering more evidence, testing alternative policies and deciding if, when, and what repairs are needed. Yet how do we know if a system's repairs are working when metrics like accuracy can be misleading? We present RagHarnessDebugger, a framework that records every repair decision in a Jev-guided harness for contract inference and audits the record to facilitate complete debugging of all decisions taken during a RAG repair journey. RagHarnessDebugger checks that each trace is intact, that each validation applies to the candidate it is counted for, and which answers the acceptance rule can reach. We tested our approach on ContractNLI non-disclosure agreements with two reader models and a replication set of separate contracts. Our debugger's audit layer exposed failure patterns that were hidden under aggregate metrics, policies acceptance cost of repairs vs regressions and policy acceptance gap of 77% and 76% of available but unrealised repairs on development and replication sets. Our audit, together with a labeled error analysis and replay mechanism, allowed following the execution path all the way to the terminal decision - uncovering where missing repairs became lost. Our revised harness improved accuracy over the initial BM25-only baseline by 3.6 points on development contracts and 3.7 points on the replication set, and even improved our full contract access reference score by 1.1 on development and 2.4 on replication set. While our work here demonstrates how the debugger can conduct decision root-cause analysis for missed repairs, the true value lies with providing this debugger for judgment visibility serving continuous improvement of RAG decisions and their explainability. Keywords: agentic RAG, AI evidence repair, harness explainability, counterfactual replay, Jev, ContractNLI

cs.AI · cs.MA · cs.SE

新论文 · v1arXiv:2611.22989

How Web Research Agents Interpret User Priorities

Robin Sharma, Parisa Suchdev

原文摘要

Web research agents gather information online and write cited reports, including financial analysis. Users can state a priority to guide this research, but a priority may not say how to judge the options, leaving the agent to choose the criteria. We examine how two such agents, built on OpenAI and Anthropic models, interpret a stated priority. In 36 runs, the agents compared Microsoft and Alphabet as businesses. Prompts asked for special attention to growth or resilience, or gave no priority. We analyzed reports, searches, returned sources and cited sources. Of the 24 reports given a priority, 21 focused on it, but their criteria varied. Under the growth priority, OpenAI reports that ranked the companies accounted for risk or predictability and favored Microsoft, while the Anthropic ones judged growth by its pace and favored Alphabet. Under the resilience priority, reports used partly different criteria, and most named Microsoft as the more resilient company. Of the search and source measures tested, only the Anthropic agent's search wording differed detectably by condition. These findings suggest evaluations should examine the criteria behind a report's judgements, not only its focus or accuracy. Interfaces could let users see and adjust those criteria.

cs.HC · cs.AI

新论文 · v1arXiv:2611.22988

Structured AI ideation yields physical strategies for epithelial barrier repair

Geonho Nah, Johyun Park, Kyung-Ah Cho, Suyoung Hwang, Yoon Hee Lim, Eun-kyung Min, Seunghyun Moon, Donghyun Koh 等 17 位作者

原文摘要

AI systems increasingly support literature analysis, candidate evaluation and hypothesis refinement, but enabling researchers to explore research directions beyond established disciplinary approaches remains challenging. We developed Abelio, an AI system designed to generate novel research directions by expanding investigators' avenues of inquiry. Abelio reformulates research objectives, uses these alternative formulations to identify relevant elements outside conventional disciplinary boundaries, and refines the resulting hypotheses. Each stage integrates divergent exploration with convergent selection through a dedicated subsystem. In quantitative evaluations, proposals generated by the Ideation Core (Abelio's hypothesis-generation core) outperformed those of four frontier large language models in scientific soundness and hypothesis diversity, and three of four in distance from existing literature. Subsystem ablation showed that each subsystem contributes to the property it targets, with the contribution of Element Selection measurable before Refinement. To investigate intestinal epithelial barrier restoration, Abelio proposed physical interventions involving membrane-tethered poly(ethylene glycol) (PEG) and cholesterol-anchored complementary DNA. Both proposals were archived before experimentation and subsequently tested independently in damaged Caco-2 monolayers. PEG preserved barrier function in a membrane-surface-dependent manner, whereas complementary DNA enabled sequence-specific restoration that was reversible on demand. These findings demonstrate how structured research ideation can identify experimentally testable approaches beyond established disciplinary conventions.

cs.AI · q-bio.QM

版本更新 · v2arXiv:2608.25846

Why ML-based cough models do not generalize: a systematic cross-dataset evaluation for tuberculosis screening

Wensi Zhang, Tomas Teijeiro, Jérôme Thevenot, David Atienza

原文摘要

Cough acoustics are promising for non-invasive tuberculosis (TB) screening, yet whether machine learning (ML) models capture disease-related acoustics or artifacts of data collection remains unresolved. We evaluated the cross-dataset generalizability of classical ML and deep learning (DL) cough-based TB classifiers across three independent datasets. Despite moderate within-dataset performance (ROC-AUC up to $0.76 \pm 0.06$), both pipelines fail to generalize, with external performance frequently below 0.6, indicating a possible limitation of the data. We further observed that audio representations are organized by recording device and dataset rather than TB status, predicted TB probability tracks country-level prevalence in CODA, and device mismatch degrades transfer while device-diverse training improves it. Additionally, a clinical-variable baseline generalizes more consistently (ROC-AUC $0.65-0.71$), indicating acquisition-specific variability in acoustics is a stronger driver of poor generalizability. High within-dataset performance is not enough. External validation is essential before cough-based TB models are clinically ready.

eess.AS · cs.AI · cs.LG · cs.SD · eess.SP

新论文 · v1arXiv:2611.22975

Large Language Models Have Not Yet Transformed Science

Ron Bekkerman, Nichole Li, Bahareh Kargar, Mark Somers

原文摘要

The emergence of large language models (LLMs) in late 2022 has raised questions about the future of science, but assessing their impact requires understanding how scientific practices were already evolving. We analyze 2.5 million full-text papers across five disciplines from 2015 to 2025, tracing trends over a broad spectrum of lexical, structural, epistemological, and social dimensions. Aside from frequency shifts in some words, most pre-2023 trends continued with limited detectable breaks. Among the notable exceptions, growth in computer science coauthor teams accelerated significantly in Asia, while apparent declines in biomedical topic diversity largely reflected standardization in article formats. Our study establishes an empirical foundation for monitoring the health of global science as LLM adoption deepens.

cs.CL · cs.AI · cs.CY · cs.DL

新论文 · v1arXiv:2611.22973

VITAL: A Framework for Validating Agentic AI in Pharmaceutical Drug Development and Clinical Trial Operations

Sandeep Burugupalli, Pedro Matos-Filipe, Andrea Hita Ardiaca, Adam Taylor, Alex Grigorenko, Natalia Jakubiak, Xin Yu, Raghav Mani 等 13 位作者

原文摘要

Current assurance approaches do not establish whether evidence is sufficient to validate, monitor and govern agentic artificial intelligence (AI) within a pharmaceutical context of use. Plausible outputs may conceal failures in retrieval, reasoning, orchestration or human-control points, where reference standards are unavailable. We developed and operationalised VITAL (Validation by Intended use, Traceability, Autonomy and Lifecycle evidence), a validation framework for agents in clinical development, through landscape analysis, regulatory and standards mapping, expert interviews and multi-stakeholder refinement. VITAL integrates context of use, controls, behavioural validation and lifecycle evidence within a risk-adjusted framework. We applied VITAL across epidemiology and treatment-landscape analysis, site selection, enrolment rescue and trial-design trade-off support to translate autonomy, reversibility, transparency and patient impact into proportionate validation activities, evidence requirements and auditable reconstruction of system behaviour. VITAL may inform evidence requirements and lifecycle accountability for higher-risk pharmaceutical uses; evaluation should assess reproducibility across assessors, organisations and system configurations.

cs.AI

新论文 · v1arXiv:2611.22971

SEAM: Computer-Use Agents Know the Harm Between Screens, and Commit It Anyway

Yuxuan Zhou, Chen Xiong, Haitong Liu, Kuofeng Gao, Yang Bai, Zhiyuan He, Tao Dai, Shu-Tao Xia

原文摘要

Safety benchmarks for computer-use agents (CUAs) place the harm either in the intent of the task or on a single screen: a malicious instruction, an injected command, a phishing page. An agent that checks the task and each screen in isolation can pass them, so they do not test whether it combines facts seen on different screens. Yet in delegated work, harm often arises exactly this way: a fact about a person on one screen, a fact about an item on another, and a legitimate task that binds them. We introduce SEAM (Safety Execution Across Multiple screens), a benchmark of such cross-screen compositional harm: 600 multi-step, cross-application OSWorld tasks in which the instruction and every single observation are benign and only their combination is harmful. Minimal-pair variants that differ in one place, scored from the server log rather than by an LLM judge, locate where an agent fails. General-purpose open-weight and flagship agents carry out the harm in up to 92% of runs, yet decline when the user states it in one sentence. Agents open the decisive pages, so exposure is not the primary bottleneck. Asked about those pages, they flag the harm, and with thinking they often write the conflict into their reasoning and confirm anyway. Safety recognition thus does not reliably translate into safe execution, and the translation depends on the information channel: the same warning stops agents far more often in the user's instruction than on screen. Proprietary models are markedly more cautious. We release SEAM with its generator and quality gates at https://yuxuan2003.github.io/SEAM/ .

cs.CR · cs.AI

新论文 · v1arXiv:2611.22967

CodeReel: Accelerating Code-to-Video Agents with Exact Frame Reuse and Blocking-Aware Scheduling

Junyi Shen, Yao Lu

原文摘要

Coding agents make videos by writing and refining programs in a build--inspect--edit loop. A finished film costs a median of $2.85 in model calls and 11 minutes, over a third of that time in builds that block the agent. We find that much of this time is avoidable, for two reasons. First, 37.5% of the frames of a session's later films were drawn before, many shifted in time or inside an edited scene. Second, half of the builds stop at a check, where the agent needs only the verdict, but which builds stop is unknown until the checks finish. We present CodeReel, a training-free execution layer behind agents' existing commands. Exact frame reuse keys each frame by its drawing commands, regardless of its position in time, and reuses each frame whose key recurs. Blocking-aware scheduling runs the checks first and starts the render alongside them, at idle priority, unless the session's recent builds make a pass unlikely. Scheduling spends CPU to answer sooner and reuse returns it; exact keys let later versions use frames rendered for a failed one as the agent thinks. We introduce EditLoop, a benchmark of replayable agent sessions whose films are checked against renders from scratch. On it, CodeReel cuts the agent's wait by 36.3% against Rebuild, a warm pipeline with parallel rendering, and by 57.9% against Sequential, today's step-by-step pipeline. On time-shifting edits its reuse renders 43.9%, close to the 43.0% whose pixels are new, while Manim-style caching renders 99.8%. With real agents, the median wait per build is 0.62x Rebuild's in Claude Code, on topics never used in development, and 0.67x in Codex CLI. The gain holds for five models and two frameworks.

cs.MM · cs.AI · cs.CV · cs.DC